SEARCH · Engineering Papers

Results for “real time experiments”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 991 records · Page 55

All sources NASA NTRS DOE OSTI

DOE OSTI2021

Exploratory analysis and performance prediction of big data transfer in High-performance Networks

Big data transfer in large-scale scientific and business applications is increasingly carried out over connections with guaranteed bandwidth provisioned in High-performance Networks (HPNs) via advance bandwidth reservation. Provisioning agents need to carefully schedule data transfer requests, compute network paths, and allocate appropriate bandwidths. Such reserved bandwidths, if not fully utilized, could be simply wasted due to the exclusive access during the approved time window, and cause extra overhead and complexity for resource management. This calls for accurate performance prediction to reserve bandwidths that match actual needs and avoid over-provisioning. We employ machine learning algorithms to predict big data transfer performance based on extensive performance measurements collected in the past several years from data transfer tests using different protocols and toolkits between various end sites on several real-life physical or emulated testbeds. We first analyze the performance patterns in response to a comprehensive list of parameters in end-host systems, network connections, and data transfer applications, which motivate the use of machine learning and also help us identify the effects of latent factors. We then propose threshold- and clustering-based methods to eliminate negative effects of latent factors in data preprocessing and build a robust performance predictor based on customized domain-oriented loss functions. The performance of the proposed methods is verified by extensive experiments using SVR and RFR as well as theoretical analysis of the general performance bound.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Exploring Flood Predictability in Taiwan through Coupled Atmospheric–Hydrological and High-Performance Hydrodynamic Models

Effective flood simulation capabilities can tremendously support early warning and disaster prevention. To examine the applicability of a fully physics-based and high-performance flood simulation and forecasting modeling framework for a flood-prone region in Taiwan, we conduct a numerical experiment that couples the Weather Research and Forecasting (WRF) Model, WRF-Hydrological modeling system (WRF-Hydro), and the Two-Dimensional Runoff Inundation Toolkit for Operational Needs (TRITON) to perform integrated rainfall, streamflow, and flood simulations. Furthermore, we first use the coupled WRF and WRF-Hydro (WWH) to predict rainfall and streamflow and then drive TRITON with the predicted streamflow hydrographs to simulate flood depth and inundation area. With the refined spatial resolution and parameterization, this framework can better predict rainfall with reasonable spatial patterns. Although WWH could overestimate the amount of rainfall in some areas, the uncertain rainfall–streamflow predictions produce reasonable flood maps able to pinpoint regions at risk of flooding. In terms of model efficiency, the graphics processing unit–based computation can yield a speed-up factor as high as ∼13 compared to the central processing unit–based computation, promoting the efficacy of the coupled modeling framework in practical real-time flood forecasting.

Coupled models↗

DOE OSTI2024

Comparing Designed Training Sets to Optimize Multivariate Regression Models for Pr, Nd, and Nitric Acid Using Spectrophotometry

Chemometric regression models were developed for the quantification of praseodymium (Pr, 0–1000 µg/mL), neodymium (Nd, 0–1000 µg/mL), and nitric acid (HNO 3 , 0.1–5 M) using spectrophotometry. Designed calibration sets were composed of 20 samples each: 10 model points and 10 lack-of-fit (LOF) points. The D-optimal designs effectively minimized the number of samples required to build models, and each design resulted in similar prediction performance, suggesting that statistical design of experiments can provide a reliable framework for selecting training set samples in three-variable systems. Partial least squares regression (PLSR) models were validated against a one-factor-at-a-time validation set composed of 125 samples (three variables, five levels). The top PLS-1 models resulted in average percent root mean square error of prediction error values of 3.5%, 1.7%, and 1.2% for Pr(III), Nd(III), and HNO 3 , respectively. Power set augmentations of the model and LOF samples were investigated to optimize the number of training set samples. PLSR models built using just required model points (10) had similar predictive capabilities as models including the LOF points (20) but with fewer samples. The number of validation samples was also varied systematically to learn how many samples are needed to validate regression models. This work addresses long-standing questions in the field of chemometrics to help make this approach amenable to the near-real-time quantification of hazardous species in remote settings.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2024

Multi-Scale Integrated Monitoring System for Enhancing Methane Emission Detection, Quantification & Prediction

This report details the progress and findings of a comprehensive study on reviewing existing solutions, identifying technology gaps, and formulating an “all-in-one” integrated strategy for developing the next-generation multiscale methane monitoring and modeling platform, conducted under grant number DE-FE0032292. Co-led by Dr. David Ebert, Dr. Binbin Weng, and Dr. Chenghao Wang at the University of Oklahoma, the project’s goal was to develop an integrated approach for building this engineering platform to detect, quantify, and mitigate methane emissions across various temporal scale, spatial scales, and sectors. The planning grant study began with an extensive review of various methane sensing and monitoring technologies and systems, surveying over 100 technology providers globally. This review revealed the prevalence of optical methods over chemical methods in commercially available sensors, with Non-Dispersive Infrared (NDIR), Tunable Diode Laser Absorption Spectroscopy (TDLAS), and Optical Gas Imaging (OGI) cameras being the most prevalent options. A trend towards more advanced optical techniques was observed, driven by increased regulatory focus and technological advancements. The technical evaluation of these sensing technologies provided crucial insights into their capabilities and limitations. The study examined emerging technologies such as Differential Absorption LiDAR (DIAL), which show promise for high-precision and long-range detection. The team then investigated the features and application bandwidth of various sensing platforms, including handheld, fixed/stationary, mobile, aerials, and spaceborne monitors. Pilot field studies were conducted to assess the capabilities of solutions for different emission scenarios. Field work with sensor deployments was conducted at three distinct site types: an oil & gas industry site, a cattle ranching operation, and a waste processing facility. The team also conducted a thorough review of methane flux inverse modeling approaches, focused on physically based methods. These approaches were categorized into simple, intermediate, and advanced methods. A realtime WRF-GHG (Weather Research and Forecasting-Greenhouse Gas) modeling system was developed and applied, incorporating multiple data sources to guide field experiments and inform methane plume detection. The project identified and analyzed numerous categories of methane data sources, including satellite measurements, ground-based sensors, and inventory databases. Key platforms examined include EDGAR, EPA GHGI, NASA TROPOMI, Carbon Mapper, and Climate TRACE, among others. The team proposed an architecture for a comprehensive methane monitoring platform. This system incorporates multi-source data acquisition, advanced data processing and assimilation, interactive visualization tools, and analytical capabilities for emissions forecasting and scenario analysis. The proposed platform aims to provide a user-friendly interface catering to various stakeholders, from researchers to policymakers. The architecture includes sophisticated data ingestion methods, a centralized data warehouse, and advanced analytical tools for data fusion and interpretation. To ensure the relevance and effectiveness of the proposed system, a comprehensive survey was conducted to gather stakeholder input on system requirements. Key findings include a strong need for integrating various data types and formats, a preference for real-time data updates and advanced visualization tools, and a demand for user-friendly interfaces catering to different expertise levels.

03 NATURAL GAS↗

DOE OSTI2023

Real-Time Monitoring of Strain Accumulation and Relief during Epitaxy of Ultrathin Co Ferrite Films with Varied Co Content

Ultrathin Co x Fe 3-x O 4 films of high structural quality and with different Co content (x = 0.6–1.2) were prepared by reactive molecular beam epitaxy on MgO(001) substrates. Epitaxy of these ferrite films is extensively monitored by means of time-resolved (operando) X-ray diffraction recorded in out-of-plane geometry to characterize the temporal evolution of the film structure. The Co ferrite films show high crystalline ordering and smooth film interfaces independent of their Co content. All Co x Fe 3-x O 4 films exhibit enhanced compressive out-of-plane strain during the early stages of growth, which partly releases with increasing film thickness. When the Co content of the ferrite films increases, the vertical-layer distances increase, accompanied by slightly increasing film roughnesses. The latter result is supported by surface-sensitive low-energy electron diffraction as well as X-ray reflectivity measurements on the final films. In contrast, the substrate–film interface roughness decreases with increasing Co content, which is confirmed with X-ray reflectivity measurements. In addition, the composition and electronic structure of the ferrite films is characterized by means of hard X-ray photoelectron spectroscopy performed after film growth. The experiments reveal the expected increasing Fe 3+ /Fe 2+ cation ratios for a higher Co content.

36 MATERIALS SCIENCE↗

DOE OSTI2021

Offline Arterial Signal Timing Optimization based on Virtual Phase Link Model - A Real-world Case Study

Conventional signal timing for arterial usually takes bottom-up approaches. Engineers optimize each individual intersection first and then coordinate them by adjusting the offsets. This work is based on the Virtual Phase-Link (VPL) model, a street traffic model designed for online traffic model predictive control, to obtain a top-down offline arterial signal timing. We Studied the Shallowford Rd. in Chattanooga, TN and found that the inconsistency in intersection capacities along the arterial could lead to some intersections becoming bottlenecks. Signal timing is a significant factor that affect the intersection capacities. We realized that the VPL-based model can guarantee the consistency in intersections along an arterial. We therefore adopted the VPL-based model and developed an offline signal timing optimization approach. The proposed timing derived from the VPL-based offline signal timing optimization showed very good results in simulation. The Chattanooga Department of Transportation adopted the optimized timing obtained from the proposed approach and gave positive feedbacks to the research team. We also collected field experiment data, which demonstrated overall energy reductions and speed improvements on some sections of the Shallowford Rd. arterial. We will continue the experiment when the COVID-19 pandemic impact subsides to have a more robust quantitative evaluation.

Wang, Qichao↗

DOE OSTI2025

Ensemble‐Based, Large‐Eddy Reconstruction of Wind Turbine Inflow in a Near‐Stationary Atmospheric Boundary Layer Through Generative Artificial Intelligence

ABSTRACT To validate the second‐by‐second dynamics of turbines in field experiments, it is necessary to accurately reconstruct the winds going into the turbine. Current time‐resolved inflow reconstruction techniques estimate wind behavior in unobserved regions using relatively simple spectral‐based models of the atmosphere. Here, we develop a technique for time‐resolved inflow reconstruction that is rooted in a large‐eddy simulation model of the atmosphere. Our “large‐eddy reconstruction” technique blends observations and atmospheric model information through a diffusion model machine learning algorithm, allowing us to generate probabilistic ensembles of reconstructions for a single 10‐min observational period. Our generated inflows can be used directly by aeroelastic codes or as inflow boundary conditions in a large‐eddy simulation. We verify the second‐by‐second reconstruction capability of our technique in three synthetic field campaigns, finding positive Pearson correlation coefficient values () between ground‐truth and reconstructed streamwise velocity, as well as smaller positive correlation coefficient values for unobserved fields (spanwise velocity, vertical velocity, and temperature). We validate our technique in three real‐world case studies by driving large‐eddy simulations with reconstructed inflows and comparing to independent inflow measurements. The reconstructions are visually similar to measurements, follow desired power spectra properties, and track second‐by‐second behavior ().

17 WIND ENERGY↗

DOE OSTI2025

Variational autoencoders for at-source data reduction and anomaly detection in high energy particle detectors

Detectors in next-generation high-energy physics experiments face several daunting requirements, such as high data rates, damaging radiation exposure, and stringent constraints on power, space, and latency. To address these challenges, machine learning in readout electronics can be leveraged for smart detector designs, enabling intelligent inference and data reduction at-source. Variational autoencoders (VAEs) offer a variety of benefits for front-end readout; an on-sensor encoder can perform efficient lossy data compression while simultaneously providing a latent space representation that can be used for anomaly detection. Results are presented from low-latency and resource-efficient VAEs for front-end data processing in a futuristic silicon pixel detector. Encoder-based data compression is found to preserve good performance of off-detector analysis while significantly reducing the off-detector data rate as compared to a similarly sized data filtering approach. Furthermore, the latent space information is found to be a useful discriminator in the context of real-time sensor defect monitoring. Together, these results highlight the multifaceted utility of autoencoder-based front-end readout schemes and motivate their consideration in future detector designs.

47 OTHER INSTRUMENTATION↗

DOE OSTI2025

Understanding EV Charging Pain Points Through Deep Learning Analysis

Current and potential electric vehicle (EV) owners express concerns about the charging infrastructure, mentioning non-functional chargers, prolonged charging times, inconvenient charger locations, long wait times, and high costs as major barriers. Addressing these issues often requires analyzing actual vehicle charging data, which is typically proprietary and inconsistent due to diverse standards and protocols. To understand and improve the EV charging experience, customer reviews are typically used to identify common customer pain points (CPPs). However, there is not a comprehensive method to map customer reviews to a standardized set of CPPs. In collaboration with the National Charging Experience (ChargeX) Consortium, this study bridges these gaps by proposing a Systematic Categorization and Analysis of Large-scale EV-charging Reviews (SCALER) framework. SCALER is an integrated, deep learning framework that segments, actively labels, analyzes, and classifies EV charging customer reviews into six CPP categories. To test its effectiveness, we used SCALER to analyze over 72,000 reviews from customers charging various EV models on different networks across the United States. SCALER achieves a classification accuracy of 92.5%, with an F1 score exceeding 85.7%. By demonstrating real-world applications of SCALER, we enhance the industry’s ability to understand and address CPPs to improve the EV charging experience.

29 - ENERGY PLANNING, POLICY AND ECONOMY↗

DOE OSTI2020

Toward quantifying capacity losses due to solid electrolyte interphase evolution in silicon thin film batteries

To understand the origins of failure and limited cycle life in lithium-ion batteries (LIBs), it is imperative to quantitatively link capacity-fading mechanisms to electrochemical and chemical processes. This is extremely challenging in real systems where capacity is lost during each cycle to both active material loss and solid electrolyte interphase (SEI) evolution, two indistinguishable contributions in traditional electrochemical measurements. Here, we have used a model system in combination with (1) precision measurements of the overall Coulombic efficiency via electrochemical experiments and (2) x-ray reflectivity measurements of the active material losses. The model system consisted of a 515 Å thick amorphous silicon (a-Si) thin film on silicon carbide in half-cell geometry using a carbonate electrolyte with LiPF6 salt. This approach allowed us to quantify the capacity lost during each cycle due to SEI evolution. Combined with electrochemical analysis, we identify SEI growth as the major contribution to capacity fading. Specifically, the continued SEI growth results in increasing overpotentials due to increased SEI resistance, and this leads to lower extent of lithiation when the cutoff voltage is reached during lithiation. Our results suggest that SEI grows more with increased time spent at low voltages where electrolyte decomposition is favored. Finally, we extracted a proportionality constant for SEI growth following a parabolic growth law. Our methodology allows for the quantitative determination of lithium-ion loss mechanisms in LIBs by separately tracking lithium ions within the active materials and the SEI and offers a powerful method of quantitatively understanding LIB loss mechanisms.

25 ENERGY STORAGE↗

DOE OSTI2025

Fostering Geothermal Machine Learning Success: Elevating Big Data Accessibility and Automated Data Standardization in the Geothermal Data Repository

The Department of Energy's (DOE's) Geothermal Data Repository (GDR) has implemented improvements to both its data lakes and its data standards and automated data pipelines. The GDR data lakes have reduced storage and compute-related barriers to using large geothermal datasets, enabling these large datasets to be accessed by anyone with a modern computer and internet access. More recently, the GDR has been working to further reduce barriers through streamlining the data intake process, educating users on the process and requirements, and helping users access data from the data lakes. These improvements have augmented the quantity of datasets the GDR is able to accept into its data lakes and have enabled users who are new to cloud tools to access these datasets more easily, overall increasing the accessibility of big geothermal data for use in machine learning and other projects. In addition, the GDR now has built-in data standards and pipelines for drilling data, geospatial data, and distributed acoustic sensing (DAS) data. These standardization efforts aim to enhance the real-world applicability of geothermal machine learning outcomes by improving the quality of training data. Specifically, through standardizing high-value datasets, the GDR is reducing project-specific data curation requirements, thus allowing more time for actual research. By automating this process, the burden of standardization is lifted from the user, ultimately increasing the availability of standardized data. This paper provides an update on recent improvements made to the GDR's data lakes and automated data pipelines, including: (1) streamlining the data lake intake process, (2) better educating users on the process and requirements through a new data lakes page, (3) adding data lake direct access links to GDR data lake submission pages, (4) implementing a DAS data pipeline to convert DAS data uploaded in SEG-Y format to a standardized hierarchical data format v5 (HDF5), (5) extending this pipeline to encompass data in the GDR data lake, (6) adding metadata requirements for geospatial data, (7) making user interface/user experience (UX) enhancements to the data pipelines' documentation pages, and (8) improving the GDR's data standards and pipelines pages to better guide users in ensuring that their data is standardized by the GDR's automated data pipelines. 2024 Geothermal Resources Council. All rights reserved.

accessibility↗

DOE OSTI2021

The mechanism driving a solid–solid phase transition in a biomacromolecular crystal

A solid-solid phase transition (SSPT) occurs between distinguishable crystalline forms. Because of its importance in application and theory in material science and condensed matter physics, SSPT has been studied most extensively in metallic alloys, inorganic salt or small organic molecular crystals, but much less so in biomacromolecular crystals. In general, the mechanism of SSPT at atomic and molecular levels is not well understood. Here, we describe the ordered molecular rearrangements in biomacromolecular crystals of the adenine riboswitch (riboA) aptamer using real-time serial crystallography and solution atomic force microscopy (AFM). The large, ligand-induced conformational changes drive the initial phase transition from the apo unit cell (AUC) to the trans unit cell 1 (TUC1). During this transition, coaxial stacking of P1 duplexes becomes the dominant packing interface, whereas P2-P2 interactions are almost completely disrupted, resulting in “floating” layers of molecules. The coupling points in TUC1 and their local conformational flexibility allow the molecules to reorganize to achieve the more densely packed and energetically favorable bound unit cell (BUC). Our study thus reveals the interplay between the conformational changes and the crystal phases—the underlying mechanism that drives the phase transition. Using polarized video microscopy (PVM) to monitor the SSPT in small crystals at high ligand concentration, we have identified the time window during which the major conformational changes take place, and simulated the in crystallo kinetics. Together, these results provide the spatiotemporal information necessary for informing time-resolved crystallography (TRX) experiments. Moreover, this study illustrates a practical approach to characterize SSPT in transparent crystals.

36 MATERIALS SCIENCE↗

DOE OSTI2022

Machine learning in nuclear materials research

Nuclear materials are often demanded to function for extended time in extreme environments, including high radiation fluxes with associated transmutations, high temperature and temperature gradients, mechanical stresses, and corrosive coolants. They also have a wide range of microstructural and chemical makeups, resulting in multifaceted and often out-of-equilibrium interactions. Machine learning (ML) is increasingly being used to tackle these complex time-dependent interactions and aid researchers in developing models and making predictions, sometimes with better accuracy than traditional modeling that focuses on one or two parameters at a time. Conventional practices of acquiring new experimental data in nuclear materials research are often slow and expensive, limiting the opportunity for data-centric ML, but new methods are changing that paradigm. Here we review high-throughput computational and experimental data approaches, especially robotic experimentation and active learning that is based on Gaussian process and Bayesian optimization. We show ML examples in structural materials (e.g., reactor pressure vessel (RPV) alloys and radiation detecting scintillating materials) and highlight new techniques of high-throughput sample preparation and characterizations, and automated radiation/environmental exposures and real-time online diagnostics. Herein, this review suggests that ML models of material constitutive relations in plasticity, damage, and even electronic and optical responses to radiation are likely to become powerful tools as they develop. Finally, we speculate on how the recent trends of using natural language processing (NLP) to aid the collection and analysis of literature data, interpretable artificial intelligence (AI), and the use of streamlined scripting, database, workflow management, and cloud computing platforms that will soon make the utilization of ML techniques as commonplace as the spreadsheet curve-fitting practices of today.

36 MATERIALS SCIENCE↗

DOE OSTI2022

First lithium experiments in HIDRA and evidence of helium retention during quasi-steady-state stellarator plasma operations

Recent experiments in Hybrid Illinois Device for Research and Applications (HIDRA) have had operational discharges between t discharge = 60 and 1000 s using electron cyclotron resonant heating (ECRH) of the plasma. This means that quasi-steady-state plasma discharges reach conditions to study long-pulse plasma material interactions (PMIs). The newly commissioned HIDRA-Material Analysis Test-stand PMI diagnostic is used to place a drop of lithium onto a heated tungsten surface, transfer the sample in-vacuo and expose it in a helium plasma. Helium is of interest as there is an open question to whether lithium will be able to remove helium ash in real fusion devices. The introduction of the W-Li sample in HIDRA resulted in evaporation of lithium into the helium plasma during a 600 s pulse and caused a reduction of over 90% in neutral pressure during the discharge. It was also observed that the plasma density and temperature increased by over 2.5 times. Using spectroscopy and a helium collisional radiative model, the peak temperature and density of the helium plasma can be monitored during the discharge. During lithium evaporation, as significant lithium ionization occurs, there is a 85% drop in the HIDRA vessel neutral pressure, despite a constant flow rate of He gas. This reduction in neutral pressure is supported by spectroscopy data with corresponding reductions in He I line intensities (587 nm, 667 nm, 706 nm, and 728 nm), as well as those of other impurities. At one point in the discharge a lithium plasma is created, as indicated by an increase in Li + emission and a complete reduction in He + emission, but the electron density jumps from ne = 3 × 10 18 m –3 to over ne = 8 × 10 18 m –3 while the core temperature stays relatively constant between Te = 16 eV and 20 eV. Once lithium has completely evaporated from the sample and the majority of the ionized lithium has diffused from the plasma to the vessel walls, pressure and spectroscopy data paired with He collisional radiative model calculations shows a re-establishment of a helium plasma in a low recycling regime. In this regime, the density drops down to ne = 2 × 10 18 m –3 and the electron temperature increases from T e = 20 eV to over T e = 50 eV indicating an increase in helium heating efficiency. This is also indicated by the He+ emission re-establishing and having a higher intensity. Here, we show the results from the first lithium campaign in HIDRA. In the presence of lithium, and in particular when lithium ions are present, the helium disappears from the plasma via an as of yet unknown complex relationship that needs to be further studied. The most likely explanation is that the lithium ions are distributed around the vessel and able to trap helium to the surface turning HIDRA into a large gettering surface. These results have potential implications on future plasma facing component design using liquid lithium for impurity and recycling control using limiters and divertors.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2023

Thermopile Energy Harvesting for Subsurface Wellbore Sensors (Final Report)

Robust in situ power harvesting underlies all efforts to enable downhole autonomous sensors for real-time and long-term monitoring of CO 2 plume movement and permeance, wellbore health, and induced seismicity. This project evaluated the potential use of downhole thermopile arrays, known as thermoelectric generators (TEGs), as power sources to charge sensors for in situ real-time, long-term data capture and transmission. Real-time downhole monitoring will enable “Big Data” techniques and machine learning, using massive amounts of continuous data from embedded sensors, to quantify short- and long-term stability and safety of enhanced oil recovery and/or commercial-scale geologic CO 2 storage. This project evaluated possible placement of the TEGs at two different wellbore locations: on the outside of the casing; or on the production tubing. TEGs convert heat flux to electrical power, and in the borehole environment, would convert heat flux into or out of the borehole into power for downhole sensors. Such heat flux would be driven by pumping of cold or hot fluids into the borehole—for instance, injecting supercritical CO 2 —creating a thermal pulse that could power the downhole sensors. Hence, wireless power generation could be accomplished with in situ TEG energy harvesting. This final report summarizes the project’s efforts that accomplished the creation of a fully operational thermopile field unit, including selection of materials, laboratory benchtop experiments and thermal-hydrologic modeling for design and optimization of the field-scale power generation test unit. Finally, the report describes the field unit that has been built and presents results of performance and survivability testing. The performance and survivability testing evaluated the following: 1) downhole power generation in response to a thermal gradient produced by pumping a heated fluid down a borehole and through the field unit; and 2) component survivability and operation at elevated temperature and pressure conditions representative of field conditions. The performance and survivability testing show that TEG arrays are viable for generating ample energy to power downhole sensors, although it is important to note that developing or connecting to sensors was beyond the scope of this project. This project’s accomplishments thus traversed from a low Technical Readiness Level (TRL) on fundamental concepts of the application and modeling to TRL-5 via testing of the fully integrated field unit for power generation in relevant environments. A fully issued United States Patent covers the wellbore power harvesting technology and applications developed by this project.

47 OTHER INSTRUMENTATION↗

DOE OSTI2026

Anomaly Detection for Online Monitoring of Thermocouple Sensors in the Advanced Test Reactor

This study explores data-driven anomaly detection methods to analyze sensor fail- ures in the Advanced Gas Reactor (AGR) nuclear fuel irradiation experiments. Specifically, we examine failures of thermocouples (TCs), which are critical for mon- itoring and controlling in-reactor temperatures during operation. Failures were pri- marily observed during abrupt power transitions and manifested as sensor drop-outs, drifts, or unexplained behavior. We applied three time-series analysis techniques— rolling mean smoothing, matrix profile, and vector auto-regression (VAR)—to de- tect anomalies in TC data prior to failure events. The rolling mean method effec- tively highlighted deviations aligned with reported failures, while the matrix profile provided partial early warning but sometimes flagged normal fluctuations during power-down periods. VAR shows potential in capturing multivariate dependencies but requires further calibration. A rare case of TC drift was also documented, which did not result in failure, underscoring the challenge of building predictive models with sparse positive examples. Our findings demonstrate that traditional statistical tools can aid anomaly detection but have limited predictive power without richer training data. We propose future directions including synthetic data generation, real- time surrogate modeling, and multi-modal feature integration. This work provides a foundation for applying robust anomaly detection frameworks to mission-critical sensor systems in experimental settings.

11 - NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

DOE OSTI2022

Computer Model Calibration with Time Series Data Using Deep Learning and Quantile Regression

Computer models play a key role in many scientific and engineering problems. One major source of uncertainty in computer model experiments is input parameter uncertainty. Computer model calibration is a formal statistical procedure to infer input parameters by combining information from model runs and observational data. The existing standard calibration framework suffers from inferential issues when the model output and observational data are high-dimensional dependent data, such as large time series, due to the difficulty in building an emulator and the nonidentifiability between effects from input parameters and data-model discrepancy. To overcome these challenges, we, in this work, propose a new calibration framework based on a deep neural network (DNN) with long short-term memory layers that directly emulates the inverse relationship between the model output and input parameters. Adopting the “learning with noise” idea, we train our DNN model to filter out the effects from data-model discrepancy on input parameter inference. We also formulate a new way to construct interval predictions for DNN using quantile regression to quantify the uncertainty in input parameter estimates. Through a simulation study and real data application with the Weather Research and Forecasting Model Hydrological modeling system (WRF-Hydro), we show our approach can yield accurate point estimates and well-calibrated interval estimates for input parameters.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2026

PowerModelsGAT-AI: Physics-Informed Graph Attention for Multi-System Power Flow With Continual Learning

Solving the alternating current power flow equations in real time is essential for secure grid operation, yet classical Newton–Raphson solvers can be slow under stressed conditions. Existing graph neural networks for power flow are typically trained on a single system and often degrade on different systems. We present PowerModelsGAT-AI, a physics-informed graph attention network that predicts bus voltages and generator injections. The model uses bus-type-aware masking to handle different bus types and balances multiple loss terms, including a power-mismatch penalty, using learned weights. We evaluate the model on 14 benchmark systems (4 to 6,470 buses) and train a unified model on 13 of these under contingency conditions with up to two branch outages, achieving an average normalized mean absolute error of 0.89% for voltage magnitudes and R 2 >0.99 for voltage angles. We also show continual learning: when adapting a base model to a new 1,354-bus system, standard fine-tuning causes severe forgetting with error increases exceeding 1000% on base systems, while our experience replay and elastic weight consolidation strategy keeps error increases below 2% and in some cases improves base-system performance. Interpretability analysis shows that learned attention weights correlate with physical branch parameters (susceptance: r=0.38 ; thermal limits: r=0.22 ), and feature importance analysis supports that the model captures established power flow relationships.

24 POWER TRANSMISSION AND DISTRIBUTION↗